Color Image Retrieval Method Fusing Principal Curvature and Color Information

Aiming at the problems of high feature dimensionality,long retrieval time,and low precision of existing algorithms in image retrieval system,an image retrieval algorithm based on the fusion of color information and texture information is proposed.Firstly,the color image is quantized and counted to obtain color histogram.Secondly,the Hessian matrix is used to extract curvature information of the color image,and the weighted joint histogram of oriented gradient based on curvature information is constructed.Then,the obtained color histogram and histogram of oriented gradient are fused to obtain the final joint histogram.And the experimental results on the Corel-1k,FTVL,and COIL-100 datasets show that the proposed algorithm has the advantages of high precision and low dimensionality.